Our overarching goal is to discover the genetic and genomic mechanisms underlying behavioral predisposition and development of addiction. Addiction remains a substantial worldwide social and economic burden despite extensive efforts to curb drug availability and use. The high heritability of drug addiction, especially for cocaine, indicates that the propensity to develop a substance use disorder after initial exposure is genetically influenced. Both human and animal studies indicate that behavioral traits such as novelty seeking and novelty preference are strongly correlated with the propensity to develop a substance use disorder, but the biological basis of this relationship is unknown. We propose to identify biological mechanisms of addiction and predisposing behavior by harnessing recent advances in mouse genetic resources, including the high-precision Diversity Outbred (DO) mouse population, validation in genetically modified mice, gene expression quantitation through RNA sequence analysis, and computational and statistical methods in systems genetics. In Aim 1 we will identify genetic mechanisms underlying predisposing novelty-related traits and drug self-administration through quantitative trait locus (QTL) analysis in a large set of DO mice. The most compelling and tractable of these will be validated in gene targeted mouse models. The intravenous drug-self administration (IVSA) paradigm, considered the gold standard for the assessment of addiction in preclinical research will enable quantification of the core features of addiction including compulsive drug use, difficulty limiting drug intake, and an extremely high motivation to take the drug. In Aim 2 we will quantify gene expression genetic variation, map expression QTLs and identify genetic correlates of predisposing behavior using RNAseq in a drug-nave subset of DO mice, and disseminate these results through widely used informatics resources. Gene expression analysis in drug-nave mice enables separation of the biological substrates of predisposition to addiction from the biological sequelae of drug exposure. By using RNA sequencing, we will be able to quantify isoforms and allelic variants in the face of genetic diversity. In Aim 3, we will address the fundamental problem of relating genetic variation in gene expression in drug-nave individuals to genetic variation in drug self-administration behavior by utilizing predisposing behavioral traits as a reference. This will be accomplished through the use of multivariate statistical methods in an approach we term reference trait genetics. This strategy makes use of a common collection of phenotypes to relate disparate and incompatible measures across two independent sets of mice. Development of this technique in the context of addiction research will extend the application of outbred populations to a wide range of applications in health and disease. Each aim will independently deliver basic research results and resources of interest to addiction biologists, including gene expression QTLs, addiction related genetic loci, and expression correlates of addiction related behavior. Synergy among the aims will reveal networks from polymorphism to addiction related behavior.